The brain storm optimization algorithm has the problems of low initial solution success rate, high computational cost and unsmooth path in solving robot path planning problem. Therefore, this paper proposed a novel brain storm optimization algorithm and its discretization scheme from the perspective of psychology. The teaching-learning idea under the herd effect enhanced the directionality of individual learning. The improved algorithm expanded the local search ratio in the later stage and improved efficiency through the step adjustment mechanism based on self-selection effect. In the discrete processing stage, it used a greedy moving search method to obtain better initial solutions. At the same time, the redefined calculation process smoothed the path in two directions. The simulation results show that the novel brain storm optimization algorithm has better performance before and after discretization, and can plan better paths in different environments. Numerical experiments verify the effectiveness of the proposed algorithm, and the application of the algorithm in the field of path planning deserves further exploration.

[2021-10-12] Accepted Paper /doc/2021.05.0213/v1

[2022-02-05] Printed Article /doc/2021.05.0213/v2